Proceedings of the 29th North American Conference on Chinese Linguistics (NACCL-29). 2017. Volume 1. Edited by Lan Zhang. University of Memphis, Memphis, TN. Pages 85-98. THE CHINA URBAN LANGUAGE SURVEY PROJECT 2003 - 2016 范德博 Marinus VAN DEN BERG Leiden University, IIAS, Emeritus In this presentation, three major hypotheses developed in the context of the China Urban Language Survey Project will be discussed. Changes in the urban language environment are mainly related to large scale migration from the country side and other places toward the newly developed and developing industrial centers in China’s southeastern provinces. The project started in 2003 in Nanjing University’s Sociolinguistic Laboratory under the guidance of professor Xu Daming, attracted researchers from various universities both in China, Hong Kong, Taiwan and Europe, and obtained funding from various sources including a major grant from the Netherlands’ Organization for Scientific Research (NWO). Progress reports were presented in yearly conferences, and major findings appeared in a 2010 book publication (van den Berg & Xu 2010). We will start this presentation with the notion “long-term accommodation”, based on work by Van den Berg in Taiwan in 1977-78 (van den Berg 1986), extend that notion to developments in mainland China, and present three major hypotheses developed in the context of the China Language Survey Project, a triglossia/diglossia hypothesis based on work in Hainan province (Tsou et al. 2010), the fundamentals of Speech Community Theory developed in studies of the language situation in the Inner-Mongolian city of Baotou (Xu 2004; 2010), and the concept of network density, developed in a study of the language situation in Beijing (Song & Zhu 2016). 1. Long-term accommodation In a questionnaire survey among National Taiwan University students in 1977, it was possible to demonstrate across generation adjustment to the national language configuration at each of the time frames involved. For grandparents of the students, who were born during the Qing Dynasty or during the beginning years of the Japanese colonial period, Minnan was the dominant language, and more so for grandmothers (71%; 74%) than for grandfathers (52%; 48%). The first of these figures giving maternal grandparents language backgrounds, and the second those of students’ paternal grandparents. These data confirm the limited options for female education during the last of the empirical dynasties, whereas Minnan-Japanese bilingualism data show the first signs of that gender gap’s closure. For maternal grandparents, the bilingualism data were 22% and 27% for VAN DEN BERG: CHINA URBAN LANGUAGE SURVEY grandmothers and grandfathers respectively, whereas Minnan-Japanese data for paternal grandparents showed a wider gap, 21% for the grandmothers, comparable to the 22% found for the maternal grandmothers, and 35% for the grandfathers, extensively higher than the figure found for the maternal grandfathers (27%). The interesting observation for the generation of parents born during the Japanese period is that a majority of them became trilingual in Minnan, Japanese, and Mandarin Chinese, called Guoyu 國 語 under the Republican government. A gender gap still existed, but both sexes had in majority become trilingual. The data for the students’ mothers was 46% and for the fathers 57%. Parents who were older or were otherwise disadvantaged in learning a new language variety did not pick-up Guoyu 國 語 , and remained bilingual Minnan-Japanese speakers, and this occurred more typically for mothers (18%) than for fathers (12%). Minnan mono-lingualism showed a similar gender difference, 20% for mothers and 9% for fathers. These latter differences most likely have a correlation with economic activity (farm labor) and income. The language situation of the students themselves gives a totally different picture. Japanese has disappeared from the language repertoire, and the best language claimed by almost all students (93%) is Mandarin Chinese (Guoyu). With an average age of 20, these students were born around 1957, and had participated in a Mandarin Chinese dominant education system, moving from elementary school, to high school, and on to university, in this case, one of the top universities in Taiwan, implying that the results are not those of the average student, but of a selection of Taiwan’s top students, who also studied in Taiwan’s political, economic, and educational center, Taipei. Keeping this in mind, these data allow the interpretation that members of a national community adjust to the national level language configuration of that community. This adjustment, using insights from interpersonal accommodation theory (Giles & Powesland 1986), we called long-term accommodation, suggesting that over-time community members, while keeping accommodating to everyday language requirements, are forced to adjust to the norms set at the national level. The result is not unexpected. It was observed for the development of Latin in the Roman Empire and related to factors such as a central government supporting economic development, making it worthwhile to acquire the language of government and education, the presence of social mobility, which helps to create multilingual areas, thereby creating the need for a lingua franca. Using this insight and turning now to mainland China, it is possible to predict that over- time the national standard language, Putonghua, will spread. The conditions for that spread, using the Taiwan and Roman data, are a well-organized education system, a language market supporting the use of Putonghua, economic development, and social mobility which will make Putonghua the aspired to lingua franca. As we will see, education got reorganized after 1980, the language market was strongly influenced by mass-migration, and the spread of Putonghua awaited the emergence of an economically developing China. 86 VAN DEN BERG: CHINA URBAN LANGUAGE SURVEY 2. Developments in Modern China Without paying attention to the pre-1980 language situation, we must conclude that the long-term accommodation hypothesis is not confirmed by developments in the 1980s. The language market in the first ten years after the start of the 1980 Open Door policy was in favor of Cantonese, the language variety of the capitalist entrepreneurs with a Hong Kong background, who invested in factories in Shenzhen and the Pearl River Delta (Zhan 1993; Guo 2004). The first wave of workers came from the surrounding country-side and were all speakers of some Cantonese dialect, and as a result oriented themselves on Cantonese rather than on Putonghua for communication. Shenzhen, the first Special Economic Zone, and the surrounding areas soon also attracted many people from other parts of China, including the Mandarin dialect area. Migrants in a new city like Shenzhen mixed home dialect languages (Hakka, Siyi (Taishanese), Swatow (Shantou), etc.) with Putonghua and Cantonese, making Putonghua the dominant language for the technical professions and for business transactions in new districts (Van den Berg 2009; Tang 2016), whereas in Cantonese speaking Guangzhou, northern and better educated migrants formed their own Putonghua based networks (Van den Berg 2010). One question that in this setting comes to the fore is how will these developments work-out for China as a whole? Various answers are possible, Putonghua will destroy the regional dialects, the dialects will stop the spread of Putonghua, or a new bilingual balance will develop. We will not discuss each of these possibilities further at this moment, but in the following we will discuss three proposals, the triglossia/diglossia hypothesis, Speech Community Theory, and network density. Each gives a somewhat different view on what is most likely the future development of the Chinese urban language market. The first of these is the strongest proposal and is the one that addresses the national language situation. It predicts a general tendency, so let’s see what that line of research has to say. 3. From triglossia to diglossia In an elementary school survey in Sanya, Hainan province, researchers observed that students, in addition to some use of the regional language variety Hainanese, mainly used the home dialect (six different language varieties), when talking to their grandparents, whereas in communication with parents, some Mandarin Chinese was introduced. When talking to each other the level of Mandarin Chinese got more extensive, reaching even higher levels when an everyday task such as shopping was involved. In public transportation, finally, Mandarin Chinese obtained its maximum use. In the latter case, we can imagine the lingua franca effect of the use of Mandarin Chinese, particularly when public transportation personnel have different backgrounds. Hainanese is still being used as lingua franca for around 20 percent of the cases, but in the remaining settings Putonghua is dominant, taking over in effect the lingua franca role of Hainanese (Tsou et al. 2010). The researchers concluded that given rapid modernization taking place in Sanya, there is a language shift taking place from the home language to, what they call, the 87 VAN DEN BERG: CHINA URBAN LANGUAGE SURVEY Supreme Language, Putonghua. They see this as a shift from a triglossia situation (home dialect-Hainanese-Putonghua) toward a diglossia setting (home language-Putonghua). And they even went one step further claiming that this development is taking place throughout China, the motivation being that a triglossic setting